Functional Connectivity Reliability in Neuroimaging Systems

Functional connectivity reliability is fundamental to interpreting patterns of synchronized neural activity derived from modalities such as functional MRI, EEG and diffusion MRI. Robust measurement underpins applications ranging from basic studies of brain organisation to the development of clinical biomarkers. Test–retest and longitudinal designs reveal that anatomical (structural) networks generally offer greater stability than functional networks, which are more sensitive to transient states, task engagement and preprocessing choices. Advances in acquisition—including multi-session and multi-modal protocols—and in analysis—such as machine learning-based metrics and hybrid resting/task frameworks—have improved reproducibility and predictive validity. Efforts to account for scanner differences, temporal structure and individual variability are converging to strengthen the reliability of functional connectomics and enhance its translational potential.

Research from Nature Portfolio

Recent longitudinal multi-session studies combining diffusion MRI, EEG and resting-state fMRI have confirmed that within-subject reproducibility exceeds between-subject reproducibility across all modalities. In EEG, alpha-band connectivity during both rest and task shows consistently high stability, whereas other frequency bands are more variable. Structural dMRI networks outperform functional fMRI networks in individual fingerprinting analyses, demonstrating superior reliability for subject identification. Among functional network measures, synchronizability and eigenvector centrality display the lowest reliability, indicating that choice of network metric should align with whether one seeks to capture stable traits or dynamic state-dependent changes.

Functional Connectivity Reliability in Neuroimaging Systems publication trend

The graph below shows the total number of articles in functional connectivity reliability in neuroimaging systems across all publications each year (not limited to Nature Index journals).

Technical terms

Functional connectivity: Statistical association or synchrony between time series of neural activity in different brain regions.

Test–retest reliability: Consistency of a measurement across repeated scanning sessions under similar conditions.

Reproducibility: Ability to obtain consistent results across studies or analytic pipelines.

Effective connectivity: Directed causal influence that one neural element exerts on another.

Structural connectivity: Anatomical white-matter pathways linking distinct brain regions.

References

1. Novel machine learning approaches for improving the reproducibility and reliability of functional and effective connectivity from functional MRI. Journal of Neural Engineering (2023).
2. Within-subject reproducibility varies in multi-modal, longitudinal brain networks. Scientific Reports (2023).
3. General functional connectivity: Shared features of resting-state and task fMRI drive reliable and heritable individual differences in functional brain networks. NeuroImage (2019).